JP2006203919A - Bidirectional ring switching control method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a bidirectional ring (BLSR) switching control method on the condition where ring switching and LP-S can not coexist but ring switching and SF-P can coexist. <P>SOLUTION: In a bidirectional ring network 1 comprising a plurality of fibers 3 and a plurality of nodes 2, in the state where (i) a first node inputs LP-S (or detects SF-P) and a second node adjacent to the first node with a span therebetween receives the LP-S (or receives the SF-P), (ii) when the second node further detects SF-R in the span, a switching request of the SF-R is transmitted from the second node to other nodes, thereby solving a problem as a whole network. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

The present invention relates to a bidirectional ring switching control method, and more particularly to a bidirectional ring (BLSR) switching control method that can be applied to a bidirectional ring network having a four-fiber configuration conforming to the North American standard SONET.
The Bidirectional Line-Switched Ring (BLSR) switching control system conforming to the SONET is realized according to the SONET Standard GR-1230-CORE Issue 4. In this method, there are two types of switching in the ring network: ring switching and span switching. In order to relieve a span in which a failure has occurred, either of the ring switching or span switching is selected. Even in this case, the protection line that forms the ring network is used.

In this case, the above switching is executed between two adjacent nodes connected to both ends of the span to be switched.
In the BLSR switching control method, information related to the switching is transmitted to each node constituting the ring network using the K1 and K2 bytes in the section overhead (SOH). The usage of the K1 and K2 bytes (K bytes) is also defined in the SONET Standard GR-1230-CORE Issue 4.

FIG. 14 is a diagram showing a configuration of a general bidirectional ring network.
In the figure, reference numeral 1 indicates a bidirectional ring network. The network 1 includes, for example, six nodes 2 (A to F) and a plurality of fibers 3 that connect adjacent nodes of these nodes to form a ring-shaped transmission line as a whole.

The plurality of fibers 3 includes a first fiber 3a in which a signal flows in a clockwise direction and a second fiber 3b in which a signal flows in a counterclockwise direction. In the drawing, a total of four fibers 3 are shown.
Each of the first fiber 3a and the second fiber 3b includes an active line (Work line) indicated by a solid line and a protection line (Protection line) indicated by a dotted line. A section between two adjacent nodes connected by the working line and the protection line is a span (4 in the figure). Each node 2 is connected on the side of Side 1 and Side 2 to the span 4 to which the node 2 is connected.

The protection line is used for relief when a failure occurs in the working line and communication on the relevant span becomes impossible. The relief method includes the ring switching and span switching described above. There is.
FIG. 15 is a diagram illustrating a state of a node when ring switching is performed. Throughout the drawings, similar components are denoted by the same reference numerals or symbols.

  In FIG. 15, the node 2 includes six nodes A to E (Node A to Node E) as in the case of FIG. 14. In this figure, for example, a failure occurs in the span 4 (denoted as 4ed in the figure) between the node E and the node D, and the ring switching (Ring Switch) is executed in the span 4ed. In the case of this figure, this ring switching is shown as an example executed by setting an FS-R (Forced Switch (ring)) by an operator, but it may be executed autonomously in the network 1 due to the occurrence of a failure. .

Here, the meanings of symbols Wed, Ped... Used in this figure will be described. It is as follows.
Wcd: Work line from node C to node D Pcd: Protection line from node C to node D Wdc: Work line from node D to node C Pdc: Protection line from node D to node C Wde: Node D to node Work line to E Pde: Protection line from node D to node E Wed: Work line from node E to node D Ped: Protection line from node E to node D Wef: Work line from node E to node F Pef : Protection line from node E to node F Wfe: Work line from node F to node E Pfe: Protection line from node F to node E First, focusing on node E, node F passes through the working line Wfe. Flow though the signal is within the node E, it is also folded back on the opposite side of the protection line Pef to the node F. This is called a ring bridge. Further, the protection line Pfe from the node F is received in the node E on the node E side instead of the working line Wde. This is referred to as ring switching. Therefore, ring switching is performed with a ring bridge.

Eventually, at node E, the working line Wed to node D is turned back to the protection line Pef to node F, and the signal taken from the working line at node D is now taken from the protection line Pfe.
On the other hand, the node D reads the FS-R from the K bytes of the signal received from the node E, and autonomously executes switching similar to the case of the node E.

  That is, the signal flowing from the node C through the working line Wcd is also returned to the protection line Pdc on the opposite side to the node C in the node D (the ring bridge (ring bridge)). Further, the protection line Pcd from the node C is connected to the working line Wdc in the node D (the ring switching (Ring Switch)).

FIG. 16 is a block diagram showing the configuration in the node. In addition, although this figure shows about one node as an example, it is the same structure also about another node.
In this figure, 11 and 15 are failure detection units, 12 and 16 are reception K byte reading units, 13 and 17 are transmission K byte writing units, and 14 is a switching control unit. In addition, reference symbols in the figure, locations where they are placed, and their meanings are as follows.

RS1a ... Side1 ... Ring switch selector RB1a ... Side1 ... Ring bridge selector SS1a ... Side1 ... Span switch selector SB1a ... Side1 ... Span bridge selector RS2a ... Side2 ... Ring switch selector RB2a ... Side2 ... Ring bridge selector SS2a ... Side2 ... span switch selector SB2a ... Side 2... Span Bridge Selector <1>, <2>, <3> and <4> in the figure represent the execution order of ring switching.

If a line failure occurs, the failure detection unit 11 detects this failure and prompts the switching control unit 14 to switch <1>. Alternatively, K bytes in the received signal are read by the received K byte reading unit 12, and the switching control unit 14 is prompted to switch <2>.
In response to this, the switching control unit 14 controls the corresponding selectors RS2a and RB1a to be in the connection state shown in the drawing. In this case, the case where only the selector RB1a is controlled is referred to as a bridge state (“Ring Bridge” in FIG. 15), and the case where not only the selector RB1d but also RS2a is controlled is the bridge switch state (“Ring Bridge” + "Ring Switch"). Note that there is no case of only the switch state.

If the failure detection unit 15 in FIG. 16 detects some line failure, the switching control unit 14 writes the APS byte into the K byte of the transmission signal and the transmission K byte writing unit 17 to the other nodes. Sequentially communicate the failure. This is communicated in the form of a switch request.
FIG. 17 is a diagram showing a K-byte format. This conforms to SONET.

The switching request is written in bits 1 to 4 of the K1 byte among the K1 byte + K2 byte constituting the K byte. Subsequent bits 5 to 8 are written with the destination node ID. In the example of FIG. 15, the ID of the node D is written.
On the other hand, in the K2 byte, the source node ID is written in bit 1 to bit 4 thereof. In the example of FIG. 15, the ID of node E is written. In subsequent bit 5, route information (short / long) is written. In FIG. 15, the path from node E → node D is short, and the path from node E → node F →... → node C → node D is long. Further, the state of the transmission source node (Idle, Bridge, Bridge & Switch (“Ring Bridge” + “Ring Switch”), etc.) is written in the following bits 6 to 8. An example of the above switching request will be described.

FIG. 18 is a list showing a specific example of the switching request. However, this conforms to SONET and is well known. Among these, LP-S and SF-P when bit 1 to bit 4 = 1111 are particularly relevant to the present invention.
The switching request information is as shown in the above list. Among them, LP-S and SF-P are indicated by the same code (1111). However, the meanings of LP-S and SF-P are different.

That is, LP-S is a command input by the user, and execution of all ring switching within the ring network 1 is prohibited. On the other hand, in SF-P, the target protection line is in the SF (Signal Fail) state, and execution of all ring switching in the ring network 1 is not prohibited.
Further, the switching to be executed in the normal ring network 1 is determined as any one switching request according to the priority order of the switching requests (see the list in FIG. 18).

  However, as an exception to such a case, there is a case where it can coexist simultaneously with a switching request having a low priority. For example, SF-P (Signal Fail (protection)) and ring switching (FS-R, SF-R, SD-R, MS-R, and EXER-R in the above list) are in the same span 4. Can coexist. That is, ring switching can be performed. Note that LP-S and ring switching cannot coexist in the ring network 1 in any case, and ring switching at this time is not executed. The following are known examples related to the present invention.

JP 2000-134244 A

  The present invention deals with a problem that occurs due to the fact that ring switching and other switching requests can coexist in the same span 4. For example, Japanese Patent Laid-Open No. 9-18419 deals with this problem. However, Japanese Patent Laid-Open No. 9-18419 refers to a problem that occurs when ring switching is canceled when ring switching and LP-S can coexist, and is different from the present invention.

That is, the present invention deals with the problem under the condition that ring switching and LP-S cannot coexist, but ring switching and SF-P can coexist.
Specifically, in the present invention, in the same Side (span side) of each node 2,
(I) Ring switching (FS-R, MS-R and EXER-R) by LP-S and command cannot be executed. (Ii) Ring switching (SF-R, SD by LP-S and line failure) -R) cannot be executed,
(iii) Ring switching (FS-R, MS-R and EXER-R) by SF-P and command is executed, or (iv) Ring switching (SF-R, SD-R by failure) with SF-P ) Is also implemented.

  In summary, in the APS byte (K byte described above) protocol used to perform bidirectional ring switching (BLSR switching) between a certain adjacent node (2) in the ring network 1, K1 indicating a switching request. LP-S and SF-P cannot be distinguished from the first 4 bits “1111” of the K1 byte in which the first 4 bits of the byte are represented by the same bit representation of “1111” for both LP-S and SF-P. Therefore, it is an object of the present invention how to discriminate these and further perform switching.

  In other words, when SF-P or LP-S and ring switching coexist, each node 2 must perform different operations for SF-P and LP-S. Since SF-P and LP-S are indicated by the same code on K bytes, only the node that detects or inputs SF-P or LP-S (for example, node E in FIG. 15) can be used only. It is impossible to know which of the two is the case. That is, correct switching cannot be executed in a node other than the node E.

That is, a node that detects SF-P or LP-S in K bytes in the received signal cannot distinguish between SF-P and LP-S, and therefore determines whether or not to perform ring switching. Can not.
Therefore, conventionally, when a node other than node E in the above example detects (1111) of the K1 byte, it interprets that either SF-P or LP-S has been received (temporarily determined). Advance the previous control. Hereinafter, a case where it is interpreted as LP-S and a case where it is interpreted as SF-P will be described with reference to a sequence diagram and the like.

FIG. 19 is a list (part 1) showing the meanings of all symbols (a1, a2,... F52) appearing in each sequence diagram.
FIG. 20 is the same list (No. 2).
Looking first at FIGS. 21-23,
FIG. 21 is a diagram showing a ring network when the K-byte receiving node operates by interpreting as LP-S reception.
FIG. 22 is an operation sequence diagram (part 1) in FIG.
FIG. 23 is the same drawing (No. 2).

Referring to FIG. 21 and FIG. 22, if a failure occurs in the protection line of the second fiber 3b in the span 4fa, the node A detects SF-P at time T1. The node A sends the first 4 bits “1111” of the K1 byte as a3 to the nodes B → C → D → E → F, and simultaneously sends it to the adjacent node F as a4.
The node F that has received this a4 flows f3 to the adjacent node A and flows f4 from the node E → D → C → B → A.

Assume that node F detects SF-R due to a line failure from node A at time T2 after a while in such a state. Here, SF-P and SF-R coexist.
In this case, even if the node F detects the SF-R, the current node F interprets the first 4 bits “1111” of the received K1 byte as LP-S, and therefore the SF-R is not executed.

In this case, the truth is that the expected result should be ring switching between nodes FA (span 4fa). However, ring switching (SF-R) due to a failure is not executed. There is one problem here.
Next, referring to FIG. 21 and FIG. 23, it is assumed that the node A is set to LP-S at time T1. The node A flows this LP-S from the node B → C → D → E → F as a5 of the first 4 bits “1111” of the K1 byte, and simultaneously flows to the adjacent node F as a6.

The node F that has received this a5 sends f3 to the node A and sends f4 to the nodes E → D → C → B → A.
Assume that node F detects SF-R due to a line failure from node A at time T2 after a while in such a state. Here, LP-S and SF-R coexist.

In this case, when the node F detects the SF-R, the current node F interprets the first 4 bits “1111” of the received K1 byte as LP-S reception. Does not execute. This is the expected result and there is no problem.
Next, a case where the first 4 bits (1111) of the K1 byte are interpreted as SF-P will be described.

FIG. 24 is a diagram showing a ring network when the K-byte receiving node operates by interpreting as SF-P reception.
FIG. 25 is an operation sequence diagram (part 1) in FIG.
FIG. 26 is the same figure (2),
FIG. 27 is the same figure (No. 3).

Referring to FIGS. 24 and 25, assume that node A inputs LP-S at time T1. In response to the input of the LP-S, the node A sends this to the node B → C → D → E → F as a11 and at the same time to the adjacent node F as a12.
The node F that has received this a12 sends f11 (RR-S) to the node A and sends f12 (LP-S) to the nodes E → D → C → B → A.

Assume that node F detects SF-R from the line from node A at time T12 after a while in such a state. Here, LP-S and SF-R coexist.
The node F that has detected SF-R sends this as f13 to the adjacent node A, and sends f14 from node E → D → C → B → A. Each node (E, D, C, B) that has received f14 performs full pass-through.

In the full pass-through state, the active line is used as it is, and the spare lines (both 3a and 3b) are also used to connect the spans on both sides of the node.
By the way, SONET also defines a communication service (PCA: Protection Channel Access) using a protection line. This is usually defined to effectively use a protection line that is not supposed to be used for communication in the ring network. If the LP-S command is input, the PCA can be used even if there is a failure. If the ring is switched, the protection line provided for the above communication service is disconnected, and the service cannot be received.

  Here, looking again at the above-mentioned full pass-through, the execution of this full pass-through causes the protection line to be in the above-mentioned connection state and incorporated into the ring network. For this reason, the communication service described above cannot be implemented in practice. In this way, the user of the communication service who expects not to be squelched due to the rescue of the working line will be hindered. This is another problem.

Next, another case will be described with reference to FIG. 24, FIG. 26, and FIG. However, in this case, since it is actually SF-P when operating assuming that it is SF-P, there is no particular problem in the conclusion.
When it is detected that there is a failure in the protection line received by node A at time T1 (SF-P), a14 is sent from node A to adjacent node F, and at the same time, a13 is sent from node A to node F.

Assume that node F detects SF-R at time T12 after a while. Here, SF-P and SF-R coexist.
The node F that has detected this SF-R sends f13 to the short path toward the adjacent node A. At the same time, f14 is caused to flow through the long path, and when it reaches node A at time T13, node A executes SF-R. Ring switching (bridge & switch) is executed here. Along with this ring switching, the node A causes a16 to flow toward the node F through the short route, and causes a15 to flow toward the node F along the long route. This a15 reaches the node F at time T13, and the node F performs ring switching. Along with this, f15 and f16 flow in both directions.

Two problems have been described above, but there is another problem.
When these switching requests change after the LP-S or SF-P switching request (the generation or recovery of the switching request), the entire ring network must perform an operation that quickly follows the change. This is because preparation is made so that the switching can be immediately executed even after the change. However, in the past, such changes have not been followed. This is another problem described above.

Therefore, the present invention
First, in the first 4 bits of the K1 byte indicating a switching request, since both LP-S and SF-P are represented by “1111”, there is no distinction and there is a ring switching request in the same span. Solves the problem that the ring switching may not be possible,
Secondly, it solves the problem that a user who receives a special communication service (PCA) while executing LP-S cannot use the protection line in use,
Thirdly, when there is a change in the switching request after the LP-S or SF-P switching request, the purpose is to solve the problem that the change cannot be followed quickly. It is.

  In order to solve the above-described problems, the present invention provides an execution node executing LP-S or SF-P, a counterpart node for switching, and another relay linked to the execution node and the counterpart node in a ring shape. By making the operation of each node different from the conventional operation, the above-described problem is avoided as a result of the entire ring network.

That is, the difference between LP-S and SF-P can be recognized. As a result, in the same span, (i) a first operation when LP-S and ring switching coexist, and (ii) a second operation when SF-P and ring switching coexist. , To make a distinction.
As a result, ring switching does not occur during execution of LP-S, and ring switching is performed while SF-P occurs. In this way, the special communication service user's request to prohibit ring switching and the user's request to carry out remedy when a failure occurs without error are simultaneously satisfied. This dramatically improves the reliability of the ring network. Furthermore, when LP-S and SF-P change, the change in the ring network can be executed in real time by following the change.

As will be explained later, according to the present invention,
First, since the K-byte protocol cannot distinguish between LP-S and SF-P, even if there is a request for ring switching in the same span, the ring switching may not be executed. Can be solved.

Secondly, it is possible to solve the problem that a user who receives a special communication service (PCA) while executing LP-S cannot use the spare line in use.
Third, when there is a change in the switching request after the LP-S or SF-P switching request, the problem that the change cannot be quickly followed can be solved.

FIG. 1 is a sequence diagram based on the present invention when LP-S and SF-R coexist in the same span.
Assume that LP-S is input to node A at time T1. A switching request indicating the first 4 bits (1111) of the K1 byte is sent to the adjacent node F as a4, and is sent from the node B → C → D → E → F as a3. The meanings of a3, a4, etc. are as shown in FIG. 19 and FIG.

Upon receiving the a4 from the node A, the node F starts transmitting a switching request indicating f4, that is, the first 4 bits “1111” of the K1 byte. F3, that is, RR-S is transmitted to the adjacent node A.
The other relay nodes (B, C, D and E) on the long path between the node A and the node F receive the above f4, respectively, and based on the information of this f4, the K byte path -Execute through (K Byte Pass-through). Unlike the above-described full pass-through, the K byte pass-through is a state in which only the received K bytes are copied and transmitted again. That is, no processing is performed on the payload portion (main signal portion) in the signal.

  After a while, at time T21, it is assumed that a failure has occurred in the working line received from node A at node F and the protection line, and this failure has been detected by node F (SF-R detection). ). Then, the node F starts to transmit the SF-R to the adjacent node A as f13, and at the same time, starts to transmit the node E → D → C → B → A as f14.

  Each of these relay nodes (E, D, C, and B) receives the above f14 and continues the K byte pass-through state based on the information of f14. The continuation of the K byte pass-through has no effect on the user of the communication service (PCA) using the protection line (solving the third problem described above).

  As described above, according to the present invention, first, in the bidirectional ring network 1 composed of a plurality of fibers 3 and a plurality of nodes 2, (i) the first node 2 (A) has LP-S (Lockout of Protection). (Span)) is set, and the second node 2 (F) adjacent to the first node 2 is receiving a switching request from the first node 2 via the fiber. Originally, (ii) when the second node 2 (F) detects a line failure in which a signal from the first node is received, the second node 2 (F) sends another node 2 (B , C, D and E), a ring switching request is transmitted.

Further, in each relay node 2 (B, C, D and E) between the first node 2 (A) and the second node 2 (F), the first node 2 (A) to the second node When the switching request having the highest priority addressed to 2 (F) is received, a K byte pass-through state in which only K bytes are passed is set.
Furthermore, in each relay node other than two adjacent nodes connected to both ends of the span to be switched, only K bytes are passed by a span switching request addressed from one of the two adjacent nodes to the other. When the K byte pass-through state is entered and a ring switching request addressed to one of the two adjacent nodes is received under this state, the K byte pass-through state is continued. It is characterized by.

FIG. 2 is a flowchart showing the switching operation in the node F of FIG.
FIG. 3 is a flowchart showing the switching operation in the node B of FIG. However, the node B is only an example, and applies to the nodes C and D.
First, in FIG.
Step S11: The node F receives LP-S (SF-P) (K bytes = 1111) from the node A (time TA).

Step S12: After that, the node F detects a failure in the reception line from the node A (SF-R detection at time T21).
Step S13: Based on the above-described SF-R detection, the node F writes SF-R in the K bytes of the transmission signal and sends it to another relay node.
As described above, the node F transmits SF-R to another node for the time being without determining whether K byte = 1111 is LP-S or SF-P. The official decision is made after a response from these other nodes. That is, the problem is solved in the entire ring network including the relay node.

An example of the switching control operation is shown in FIG.
Step S21: The node B receives a3 (LP-S) addressed to the node F from the node A (time T11).
Step S22: The node B receives f4 (LP-S) addressed to the node A from the node F (time T12).

Step S23: The node B enters the K byte pass-through state upon receiving the above a3 and f4.
Step S24: Thereafter, the node B receives f14 (SF-R) addressed to the node A from the node F (time T22).
Step S25: The node B compares the received SF-P (LP-S) with SF-R. In this case, since SF-P (LP-S) has a higher priority than SF-R (see FIG. 18), the above K byte pass-through state is continued. Conventionally, the full pass-through state is established at this time.

The above case is a case where LP-S and SF-R coexist. Next, a case where SF-P and SF-R coexist will be described.
FIG. 4 is a sequence diagram (No. 1) based on the present invention when SF-P and SF-R coexist in the same span.
FIG. 5 is the same figure (the 2).

First, referring to FIG.
It is assumed that at time T1, the node A detects SF-P due to a reception protection line failure from the node F. The node A executes this SF-P, and sends a switching request for the K1 byte “1111” as a14 to the adjacent node F, and also as a13 from node B → C → D → E → F. The meanings of a13, a14, etc. are as shown in FIGS.

When the node F receives the switching request (a14) of the K1 byte “1111” from the node A, the node F starts to transmit the switching request (f12) of the K1 byte “1111” as a response thereto. F11, that is, RR-S is transmitted to the adjacent node A.
The other relay nodes (B, C, D and E) on the long path between the node A and the node F receive the above f12, respectively, and based on the information of this f12, the K byte path -Execute through (K Byte Pass-through).

  After a while, it is assumed that at time T31, the node F has detected SF-R due to a reception working and protection line failure from the node A (SF-R detection). Then, the node F executes the SF-R, starts to transmit this SF-R to the adjacent node A as f13, and simultaneously starts to transmit to the nodes E → D → C → B → A as f14.

  Each of these relay nodes (E, D, C, and B) receives the above f14 and continues the K byte pass-through state based on the information of f14. The continuation of the K byte pass-through has no effect on the user of the communication service (PCA) using the protection line (solving the third problem described above).

  As described above, according to the present invention, in the bidirectional ring network 1 including the plurality of fibers 3 and the plurality of nodes 2, (i) the first node 2 (A) detects the reception standby line failure from the second node. Detecting and the second node 2 (F) adjacent to the first node 2 (A) is receiving a switching request from the first node via the fiber (Ii) When the second node 2 (F) detects a reception line failure from the first node, the second node 2 (F) transmits a ring switching request to the other nodes. It is characterized by doing so.

In each relay node 2 (B, C, D and E) between the first node 2 (A) and the second node 2 (F), the first node 2 (A) to the second node When the switching request having the highest priority addressed to 2 (F) is received, a K byte pass-through state in which only K bytes are passed is set.
Furthermore, in each relay node other than two adjacent nodes connected to both ends of the span to be switched, only K bytes are allowed to pass by a span switching request addressed from one of the two adjacent nodes to the other. When the byte switch-through state is entered and a ring switching request addressed to one of the two adjacent nodes is received under this state, the K-byte pass-through state is continued. It is a feature.

Next, the time after time T32 in FIG. 4 will be described with reference to FIG.
At time T32, the node A receives f14 from the node F through the long path (lower end in FIG. 4). Thereby, the node A transmits SF-R as a31 on the long path. At the same time, a32 is transmitted to the adjacent node F.
That is, in the other node 2 (A) that has received the ring switching request from one node 2 (F) of two adjacent nodes connected to both ends of the span to be switched, the other node 2 (A) If LP-S or SF-P has been received before that, only the ring switching request corresponding to the ring switching request is transmitted, and the idle state is continued without executing the ring switching. .

Conventionally, the bridge switch (ring switching) is performed without transmitting the SF-R.
In the above case, the relay nodes (B, C, D, and E) on the long path between the node A and the node F receive the SF-R (a31), and thus the full pass-through. It becomes a state. At time T33, the node F receives the SF-R (a31), and thereby executes a bridge switch (Ring Br & Sw in the figure).

Accordingly, the node F transmits f31 to the adjacent node A side, while transmitting f32 (Br & Sw) toward the node A through the long path.
The node A that has received this f32 executes ring switching (Ring Br & Sw in the figure) at the node A. Further, the node A transmits a34 and a33 toward the node F.

FIG. 6 is a flowchart showing an operation example of the relay node in FIGS. 4 and 5. A description will be given by taking the node B as an example of the relay node.
Step S31: The node B in FIG. 4 receives LP-S addressed from the node A to the node F as a13 (time T11).
Step S32: The node B enters a K byte pass-through state.

Step S33: The node B receives LP-S on the long route transmitted from the node F to the node A as f12 (time T12).
Thereafter, the node F detects SF-R at time T31 and transmits f13 and f14.
Step S34: The node B receives SF-R transmitted from the node F to the node A as f14 (time T31 ′).

  Step S35: When the above f14 is received in step S34, the above-described priorities of SF-P (LP-S) and SF-R received by the node B are compared. Since SF-P (LP-S) has a higher priority than SF-R, K-byte pass-through is continued. Conventionally, the full pass-through state has been achieved at this time.

Step S36: In FIG. 5, the node B receives SF-R addressed to the node F from the node A as a31 at time T32 ′.
Step S37: Since the requests received by both Sides (Side 1 and Side 2) of the Node B are both SF-R, full pass-through is executed for the first time here.

Next, a description will be given of a case where a change in LP-S and SF-P, that is, occurrence or recovery of one node in the ring network 1 occurs.
FIG. 7 is a sequence diagram (part 1) when LP-S and SF-P change in a node and coexist with SF-R.
FIG. 8 is the same figure (the 2).

FIG. 7 is a continuation of the sequence shown in FIG. Returning to FIG. 1, K bytes transmitted and received at each node without any trouble in the ring network 1 are a1, a2, b1, b2, c1, c2, d1 shown in the upper part of FIG. , D2, e1, e2, f1 and f2.
Thereafter, the LP-S described above is input at time T1, and the node A executes this LP-S in the span 4 between the node A and the node F.

At this time, the node F receives a4 from the node A, and transmits f3 and f4 in response to this (described above).
Further, at time T21, the node F detects SF-R from the reception working and protection lines from node A, and executes this SF-R (described above).
Thereafter, it is assumed that SF-P is detected at node A at time T42 in FIG. However, since this SF-P has the same code (K = 1111) as LP-S, the K bytes transmitted from the node A to other nodes do not change.

Under this state, assuming that LP-S is released at node A at subsequent T43, node A executes SF-R based on the K bytes received from node F. This SF-R is transmitted from the node A to other nodes as a41 and a42.
The node F receives the a41 at time T44 and performs ring switching (“Ring Br & Sw” in the figure). Along with this, f41 and f42 are transmitted from the node F, and f42 reaches the node A through the long path. In response to this, the node A performs ring switching (Ring Br & Sw), and transmits a43 and a44 accordingly.

From this state, time T45 in FIG. 8 is reached. Assume that LP-S is input to node A again at time T45. In response to this, the node A starts to transmit the LP-S to the node F as a45 and a46.
At each relay node (B, C, D, and E) that receives the a45, the contents of the f42 (second half of FIG. 7) received in the opposite direction, that is, SF-R / of FIG. 19 and FIG. Looking at node A (transmission destination) / node F (transmission source) / long / Br & Sw, since the status in the K2 byte is Br & Sw, the full pass-through state (each Full Pass in FIG. 8) -Through).

  The node F receives the a45 relayed at time T45 'and cancels the ring switching ("Ring Br & Sw cancel execution" in FIG. 8). Then, an idle state (“Idle state” in the figure) is entered. Accordingly, the node F transmits SF-R as f43 and f44 to the node A through the adjacent node A and the long route.

The f44 is received by each relay node (E, D, C, and B) through the long route, and each relay node that receives the f44 executes K byte pass-through ("K Byte Pass- in the figure"). through).
The operation in FIGS. 7 and 8 will be further described. Each of the relay nodes 2 (B, B) other than the two adjacent nodes 2 (A) and 2 (F) connected to both ends of the span 4 to be switched. C, D, and E) are connected to the full path − by the ring switching request sent from one node 2 (A) of two adjacent nodes 2 (A) and 2 (F) to the other node 2 (F). When a span switching request addressed to the other node 2 (F) is received from one node 2 (A) under that state, each priority of the received ring switching request and span switching request is received. K byte pass-through state in which only K bytes are passed when the priority is compared and the priority order of the span switching request is higher and the status code related to the ring switching request is not a ring bridge or a ring switch. So as to.

In the above-described embodiments of FIG. 1 and FIG. 4, it has been described that the switching is performed by transmitting the SF-R when the SF-R is detected. However, in addition to this embodiment, the EXER-R is transmitted. It is also possible to execute switching.
FIG. 9 is a diagram (No. 1) showing an embodiment in which switching is realized by transmitting EXER-R when SF-R is detected,
FIG. 10 is the same figure (the 2).

First, in FIG. 9, it is assumed that node A detects SF-P at time T1. Thereby, the node A transmits SF-P as a13 and a14 to the node F through the long route and the short route.
The node F that has received the SF-P (a14) transmits f11 and f12 as illustrated.

  Thereafter, at time T51, it is assumed that the node F detects SF-R in the reception working and protection lines from the node A. As a result, the node F attempts to perform ring switching, but the K byte information transmitted to the other node is EXER-R. The EXER-R can be considered as a switching test signal. That is, by sending this EXER-R through the ring network, it is possible to confirm whether or not the partner node is in a switchable state.

  The EXER-R is transmitted from the node F to the other nodes as f51 and f52. f51 is transmitted to the adjacent node A, and f52 is transmitted to each relay node (E, D, C, and B) through the long path and to the destination node A. Since each relay node that has received the f52 receives LP-S and EXER-R, the relay node enters a K byte pass-through ("K Byte Pass-through" in FIG. 9) state.

Next, referring to FIG. 10, the node A that has received f52 at time T52 in the lower end of FIG. 9 executes SF-R at time T53 in FIG. 10, and uses this SF-R as a31 and a32 through the long path. The data is transmitted from node B → C → D → E → F and transmitted to adjacent node A.
The a31 is received by the node F at time T54, and the node F that has received the a31 performs ring switching ("Ring Br & Sw execution" in the figure).

Accordingly, the node F transmits SF-R as f31 and f32 toward the node A as illustrated.
When node A receives f32 at time T54 ′, it executes “Ring Br & Sw” shown in the figure, and completes ring switching.
In the sequence described above, the following operations are shown in the flowchart with attention paid to the switching control at the node F and the switching control at the node A, respectively.

FIG. 11 is a flowchart showing the switching control operation at the node F.
FIG. 12 is a flowchart showing the operation of switching control at the node A.
First, in FIG.
Step S41: The node F receives SF-P (LP-S) during times T1 to T51 in FIG.

Step S42: The node F detects SF-R on the reception line from the node A at time T51.
Step S43: Although the node F detects the SF-R, but is in the LP-S reception state, it transmits EXER-R as f51 and f52 to other nodes. Receiving this, node A transmits a31 and a32 after time T53.

One of the features of the present embodiment is that the EXER-R is transmitted here.
Step S44: Thereafter, the node F receives the a31 from the node A at T54. The reception of SF-R.
Step S45: Subsequently, the node F executes “Ring Bridge & Switch”, and sends SF-R as f31 and f32 to other nodes.

Next, referring to FIG.
Step S51: The node A detects SF-P on the reception protection line at time T1 in FIG.
Step S52: The node A transmits the SF-P as a13 and a14 to other nodes.

At time T51 in FIG. 9, the node F transmits f51 and f52 upon detection of SF-R. These f51 and f52 are the EXER-R described above.
Step S53: The node A receives EXER-R from the node F, that is, the above f52 at time T52.
Step S54: In FIG. 10, the node A determines that it is a ring switching request from the node A (after time T53), and sends SF-R as a31 and a32 to other nodes.

The node F that has received the a31 executes ring switching and transmits f31 and f32 (after time T54).
Step S55: The node A receives the f32 and realizes ring switching at time T54 ′.
9 to 12 described above, according to FIG. 9, in a bidirectional ring network composed of a plurality of fibers 3 and a plurality of nodes 2, (i) the first node is LP-S (Lockout of (Protection (span)) and the second node adjacent to the first node across the span receives the LP-S through the fiber. ii) When the second node further detects SF-R (signal Fail (ring)) in the span, the second node transmits an EXER-R (Excerser (ring)) to another node. A switching request is transmitted.

  Alternatively, (i) the first node detects SF-P (Signal Fail (Protection)), and the second node adjacent to the first node across the span passes the SF through the fiber. (Ii) When the second node further detects SF-R (signal Fail (ring)) in the span under the condition of receiving -P, from the second node, An EXER-R (Excerser (ring)) switching request is transmitted to the node.

  Further, according to FIG. 10, in the bidirectional ring network 1 composed of a plurality of fibers 3 and a plurality of nodes 2, of the first and second nodes adjacent to each other across the span to be switched. When the first node receives an EXER-R (Exerciseer (ring)) switching request from the second node while executing an SF-P (Signal Fail (Protection)) switching request, the first node The node transmits a switching request of SF-R (signal Fail (ring)).

  In the last embodiment, LP-S and SF-P are distinguished in the same K byte. That is, at least LP-S (Lockout of Protection (span)) switching request and SF-P (Signal Fail (Protection)) switching request using K bytes from a node connected to the span to be switched. Alternatively, LP-S and SF-P are identified using unused bit areas in K bytes.

Referring again to FIG. 17, focus on bits 6-8 of the K2 byte.
FIG. 13 is a diagram illustrating an example in which LP-S and SF-P are identified using the K2 byte.
According to GR1230-CORE, 100 and 101 of the K2 byte are “Reserved for future use”, which is newly defined as shown in FIG. 13, and LP-S is clearly distinguished from SF-P. It can be so.

In actual operation, when the node recognizes LP-S, the operation of FIG. 23 may be realized. On the other hand, when it is recognized as SF-P, the operation according to FIGS. 26 and 27 is performed. Thereby, a normal switching operation can be realized.
The embodiment of the present invention described above is as follows.
(Appendix 1) In a bidirectional ring network composed of a plurality of fibers and a plurality of nodes,
(I) LP-S (Lockout of Protection (span)) is input to the first node, and a second node adjacent to the first node is connected to the first node via the fiber. Under the state of receiving the switching request from
(Ii) when the second node detects a failure of a line that receives a signal from the first node;
A bidirectional ring switching control method, wherein a ring switching request is transmitted from the second node to another node.
(Supplementary Note 2) In a bidirectional ring network composed of a plurality of fibers and a plurality of nodes,
(I) the first node detects that the reception protection line from the adjacent second node is faulty, and the second node receives the failure from the first node via the fiber; Under the condition that the switch request is received,
(Ii) when the second node detects a failure of a line received from the first node;
A bidirectional ring switching control method, wherein a ring switching request is transmitted from the second node to another node.

(Supplementary note 3) When each relay node between the first node and the second node receives a switching request with the highest priority addressed to the second node from the first node, 2. The bidirectional ring switching control method according to appendix 1, wherein a K byte pass-through state in which only K bytes pass is set.
(Supplementary note 4) When each relay node between the first node and the second node receives a switching request with the highest priority addressed to the second node from the first node, 3. The bidirectional ring switching control method according to appendix 2, wherein a K byte pass-through state in which only K bytes pass is set.
(Supplementary Note 5) In a bidirectional ring network composed of a plurality of fibers and a plurality of nodes,
In each relay node other than two adjacent nodes connected to both ends of the span to be switched,
A span switching request addressed from one of the two adjacent nodes to the other results in a K byte pass-through state in which only K bytes pass, and under that state, from one of the two adjacent nodes to the other. A bidirectional ring switching control method characterized by continuing the K-byte pass-through state when receiving a ring switching request addressed thereto.

(Appendix 6) In a bidirectional ring network composed of a plurality of fibers and a plurality of nodes,
In the other node that has received the ring switching request from one of the two adjacent nodes connected to both ends of the span to be switched, the other node has previously received LP-S or SF-P. In such a case, a bidirectional ring switching control method is characterized in that the idle state is continued only by transmitting a ring switching request corresponding to the ring switching request without executing the ring switching.
(Appendix 7) In a bidirectional ring network composed of a plurality of fibers and a plurality of nodes,
Each relay node other than two adjacent nodes connected to both ends of the span to be switched is
A ring switching request sent from one of the two adjacent nodes to the other node causes a full pass-through state, and under that state, the span switching from the one node to the other node is performed. When the request is received, the priority of the received ring switching request and the span switching request is compared, and the priority of the span switching request is higher, and the status code related to the ring switching request is the ring A bidirectional ring switching control method characterized in that, if it is not a bridge and a ring switch, it enters a K byte pass-through state in which only K bytes pass.
(Supplementary note 8) In a bidirectional ring network composed of a plurality of fibers and a plurality of nodes,
(I) LP-S (Lockout of Protection (span)) is input to the first node, and a second node adjacent to the first node is connected to the first node via the fiber. Under the state of receiving the switching request from
(Ii) when the second node detects a failure of a line received from the first node;
A bidirectional ring switching control method, wherein a ring switching request is transmitted from the second node to another node.

(Supplementary note 9) In a bidirectional ring network composed of a plurality of fibers and a plurality of nodes,
(I) The first node detects a protection line failure received from the adjacent second node, and the second node receives the switching request from the first node via the fiber. Under the condition
(Ii) when the second node detects a failure of a reception line from the first node;
A bidirectional ring switching control method, wherein a ring switching request is transmitted from the second node to another node.
(Supplementary Note 10) In a bidirectional ring network composed of a plurality of fibers and a plurality of nodes,
While executing an SF-P (Signal Fail (Protection)) switching request at the first node of the first and second nodes adjacent to each other across the span to be switched, When an EXER-R (Excersizer (ring)) switching request is received from the node 2,
The first node transmits a switching request of SF-R (signal Fail (ring)).
(Supplementary Note 11) In a bidirectional ring network composed of a plurality of fibers and a plurality of nodes,
From a node connected to the span to be switched, using the K byte, at least an LP-S (Lockout of Protection (span)) switching request and an SF-P (Signal Fail (Protection)) switching request, A bidirectional ring switching control method characterized in that when transmitting alternatively, the LP-S and the SF-P are identified using an unused bit area in the K bytes.

As described above, according to the present invention,
First, since the K-byte protocol cannot distinguish between LP-S and SF-P, even if there is a request for ring switching in the same span, the ring switching may not be executed. Can be solved.

Secondly, it is possible to solve the problem that a user who receives a special communication service (PCA) while executing LP-S cannot use the spare line in use.
Third, when there is a change in the switching request after the LP-S or SF-P switching request, the problem that the change cannot be quickly followed can be solved.

It is a sequence diagram based on this invention in case LP-S and SF-R coexist in the same span. It is a flowchart which shows the switching operation in the node F of FIG. It is a flowchart which shows the switching operation in the node B of FIG. It is the sequence diagram (the 1) based on this invention in case SF-P and SF-R coexist in the same span. It is the sequence diagram based on this invention in case SF-P and SF-R coexist in the same span (the 2). 6 is a flowchart illustrating an operation example of a relay node in FIGS. 4 and 5. FIG. 6 is a sequence diagram (part 1) when LP-S and SF-P change in a node and coexist with SF-R. FIG. 11 is a sequence diagram (part 2) when LP-S and SF-P change in a node and coexist with SF-R. It is FIG. (1) which shows the Example which implement | achieves switching by transmitting EXER-R when SF-R is detected. It is FIG. (2) which shows the Example which implement | achieves switching by transmitting EXER-R when SF-R is detected. 4 is a flowchart showing an operation of switching control at a node F. 5 is a flowchart showing an operation of switching control at a node A. It is a figure which shows an example which identifies LP-S and SF-P using K2 byte. It is a figure which shows the structure of a general bidirectional ring network. It is a figure which shows the mode of the node at the time of ring switching. It is a block diagram which shows the structure in a node. It is a figure showing the format of K byte. It is a list | wrist which shows the specific example of a switching request | requirement. It is the list (the 1) which shows the meaning of all the symbols (a1, a2 ... f52) which appear in each sequence diagram. It is the list | wrist (the 2) which shows the meaning of all the symbols (a1, a2 ... f52) which appear in each sequence diagram. It is a figure which shows a ring network in case a K byte receiving node operates by interpreting as LP-S reception. FIG. 22 is an operation sequence diagram (part 1) in FIG. 21; FIG. 22 is a second operation sequence diagram in FIG. 21. It is a figure which shows a ring network in case a K byte receiving node operates by interpreting as SF-P reception. FIG. 25 is an operation sequence diagram (part 1) in FIG. 24; FIG. 25 is an operation sequence diagram (part 2) in FIG. 24. FIG. 25 is an operation sequence diagram (part 3) in FIG. 24;

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Bidirectional ring network 2 Node 3 Multiple fibers 4 Span 11 Failure detection part 12 Reception K byte reading part 13 Transmission K byte writing part 14 Switching control part 15 Failure detection part 16 Reception K byte reading part 17 Transmission K byte writing Part

Claims (5)

In a bidirectional ring network consisting of multiple fibers and multiple nodes,
(I) LP-S (Lockout of Protection (span)) is input to the first node, and a second node adjacent to the first node is connected to the first node via the fiber. Under the state of receiving the switching request from
(Ii) when the second node detects a failure of a line that receives a signal from the first node;
A bidirectional ring switching control method, wherein a ring switching request is transmitted from the second node to another node. In a bidirectional ring network consisting of multiple fibers and multiple nodes,
(I) the first node detects that the reception protection line from the adjacent second node is faulty, and the second node receives the failure from the first node via the fiber; Under the condition that the switch request is received,
(Ii) when the second node detects a failure of a line received from the first node;
A bidirectional ring switching control method, wherein a ring switching request is transmitted from the second node to another node. In a bidirectional ring network consisting of multiple fibers and multiple nodes,
In each relay node other than two adjacent nodes connected to both ends of the span to be switched,
A span switching request addressed from one of the two adjacent nodes to the other results in a K byte pass-through state in which only K bytes pass, and under that state, from one of the two adjacent nodes to the other. A bidirectional ring switching control method characterized by continuing the K-byte pass-through state when receiving a ring switching request addressed thereto. In a bidirectional ring network consisting of multiple fibers and multiple nodes,
(I) LP-S (Lockout of Protection (span)) is input to the first node, and a second node adjacent to the first node is connected to the first node via the fiber. Under the state of receiving the switching request from
(Ii) when the second node detects a failure of a line received from the first node;
A bidirectional ring switching control method, wherein a ring switching request is transmitted from the second node to another node. In a bidirectional ring network consisting of multiple fibers and multiple nodes,
(I) The first node detects a protection line failure received from the adjacent second node, and the second node receives the switching request from the first node via the fiber. Under the condition
(Ii) when the second node detects a failure of a reception line from the first node;
A bidirectional ring switching control method, wherein a ring switching request is transmitted from the second node to another node.

Images